Stock Forecast Algorithms: An Overview

dr roitmanCo-Founder & CTO of I Know First Ltd. With over 35 years of research in AI and machine learning. Dr. Roitman earned a Ph.D  from the Weizmann Institute of Science A stock forecast relies on a series of accumulative events which determine the rise and fall of a stock. While stock forecast is not always entirely accurate, there seems to be a great deal of accuracy among many of the most popular stock market

Read More

Google Stock Forecast: The Algorithmic Perspective

Google (NASDAQ:GOOG) (NASDAQ:GOOGL) reported a fantastic second quarter and Wall Street reciprocated with shares jumping in the following days. Results surged way beyond last year's second quarter report and the company successfully beat most analysts' revenue expectations. As this jump in value is certainly justified, there are still some concerns that we should not forget including Google's declining cost-per-click, and the Facebook's (NASDAQ:FB) growing share of the mobile advertising market. On the other hand, Google has future prospects that should have investors excited about the company's future. From utilizing YouTube ads to pushing the innovation envelope with Google Glass, the long-term prospects are certainly interesting and if executed correctly, shareholders will be rewarded.

The company's performance this past quarter was impressive on most figures including increased revenue that was up 22%, from $12.67 billion in the second quarter of 2013 to $15.96 billion in 2Q14. Google's Non-GAAP net income came in at $4.18 billion in the second quarter, beating $3.36 billion in the same quarter last year. "Other revenues" jumped significantly this past quarter by an impressive 53% from last year's 1.05 billion to 1.6 billion. This segment includes app download revenue, including a large array of games, which is an area of focus for Google. Figure 1 shows how Google has been able to successfully increase revenue since the first quarter of 2008. (click to enlarge) While there were many strong points reported within this earnings report, Google's declining cost-per-click (CPC) for Google's sites were down 7% and average CPC declined 6% since last year. Pricing pressure has increased and will continue to increase due to Facebook's strong mobile ad growth.

Read More

Did You Know? Using Big Data For Forecasting

Find the Next Big Trend

We are active in patent mining and forecasting emerging technology trends in the fields of pharmaceuticals, energy storage, and superconductor technology. We monitor patents filed and track the evolution of concepts. We predict inventions!

We track the number of patents and their quality in different patent classes with time and analyze the data, using algorithms that model the interaction between competing approaches, and project the future trajectory of the given approach in the multidimensional space of other approaches.

Extracting Meaning from Big Data

We are constantly

Read More

Stock Wave: What You Need To Know

dr roitmanCo-Founder & CTO of I Know First Ltd. With over 35 years of research in AI and machine learning. Dr. Roitman earned a Ph.D  from the Weizmann Institute of Science Several tools are popularly applied in the world of stock trading, such as algorithms, market forecasts, and detailed analyses of specific stocks. Regardless of the theory or the practice one takes up, the very basics of stock trading are split into three distinct categories: Random walk, stock trends and stock wave.

Read More

Stock Market Forecast: The ‘Gut Feeling’ Challenge (Seeking Alpha Article)

S&P 500 Forecast Using Algorithms: A New Method.

By Lipa Roitman Ph.D.
May 12 2013

It is well known that the market as a whole is one living interacting system. What goes on in one industry sector has implications for the rest of the economy.  That is why most stocks move up and down together in waves.

The I Know First algorithms predict hundreds of stocks, indices, currencies, and commodities.  The stocks and stock indices among them are a snapshot of the whole economy, representing the major industry sectors. The forecast is published as a color coded heat map table with green cells indicating up signal, the red ones a down signal. The table helps to quickly get an idea of predictions for the whole assembly of stocks. It would be interesting to see whether a simple measure of the average signal for the whole group of stocks could predict the broader market movement. The chart below (Fig.1) shows such statistics for the I Know First algorithms forecasts over a period of the last 18 months. In lieu of the signals for individual stock, a combined signal of all stocks in the system was calculated for each of the six time ranges. The positive or negative (up or down) signals of the forecast were added to the actual last known price at the time of forecast. Thus, when the signal line is above the actual line, it means buy, if below, then sell. Each point on this chart was taken from the actual daily forecast published in the morning before the next market open. There are six times range forecasts that the system outputs, from three days to one year ahead. One can see from the chart that the new way of using I Know First algorithms correctly predicted the two major up trends, September 2011 to July 2012, and November 2012 till today. The July-November 2012 signals were too weak to be actionable.   See: Fig. 1. S&P500 chart (thick blue line) vs. combined signals of all stocks in the system. S&P 500 forecast The history of the I Know First published forecasts. The S&P 500 is up by 7.7 percent in the last 3 months. For the year, the benchmark S&P 500 is up a stunning 14.6 percent. In the October 16th 2012 Seeking Alpha article “2 S&P 500 Forecasts For 2012-13” we predicted that the markets would fall in November. And they fell indeed. Shortly before the November 14 lows the I Know First algorithms have signaled a broad market upturn. We have reported on the impending stock market up-trend as early as November 8, 2012. In the post-elections Seeking Alpha article we wrote: “…the long-term forecast for a year ahead is improving, meaning the S&P 500 will be mostly higher next year than what where it’s at today. “ In the next, December 27 article “Fiscal Cliff or Not, the Economy is Recovering” we have emphasized a number of positive signals based on the I Know First algorithms forecasts, and concluded: “The economy is recovering steadily”. On March 17, in the article “Stock Market Forecast: Two Recovery Signals Based on Algorithms” and in the following continuation articles “…Part 2 (Australia, New Zealand, Asia)” and “…Part 3 (Latin America Indices)” we showed a picture of broad and strong recovery signs all over the world based on many index forecasts charts.. Conclusion: The stocks and indices in the I Know First system are a good representation of a broader market. The forecast for the plurality of stocks in the I Know First system can serve as a proxy for the S&P500 forecast.


How Can We Predict the Financial Markets By Using Algorithms? university talk by Lipa Roitman PhD (co-founder of I Know First)